We are developing human parainfluenza viruses (HPIVs) as vaccine vectors for human use against highly pathogenic emerging viruses, with a present focus on HPIV serotypes 1 and 3. This strategy takes advantage of the natural respiratory tract tropism of HPIVs to provide respiratory administration and stimulate strong systemic immune responses as well as local mucosal immunity that is important for restricting pathogens that infect and are spread via the respiratory tract and conjunctiva. HPIV3 vector: We previously constructed a 1st generation construct called HPIV3/EboGP, in which the complete HPIV3 genome was modified by the addition of the EBOV GP gene in the third gene position, between the HPIV3 P and M genes. EBOV GP is the sole EBOV virion surface protein and is the sole EBOV neutralization antigen and major protective antigen. The EBOV GP gene was engineered to have the appropriate HPIV3 transcription signals for it to be expressed as a separate mRNA by the HPIV3 polymerase. HPIV3/EboGP was substantially immunogenic and protective when given to non-human primates by combined intranasal (IN) and intratracheal (IT) administration, even in animals previously infected with HPIV3. However, immunogenicity depended on IT delivery of vaccine: IN delivery alone was insufficient. This suggested that vector expression beyond the upper respiratory tract was necessary for immunogenicity. IT delivery in humans would not be feasible, but might be substituted by aerosol delivery. The aerosol route has been used by others to immunize humans in large vaccine trials, such as with the measles virus vaccine, and suitable delivery devices exist. We explored delivery of the HPIV3/EboGP construct by the aerosol route in rhesus macaques. The aerosol route was generally more immunogenic and protective than the combined IN/IT route. This included generally higher serum and mucosal EBOV-specific IgG, IgA, and neutralizing antibody titers, as well as EBOV-specific cellular responses in the lungs, including polyfunctional CD8+ T cells and CD4+ T helper cells that were predominately Th1. In addition, the HPIV3/EboGP vaccine produced more robust cell-mediated and humoral immune responses than an alphavirus vaccine delivered parenterally in parallel. One aerosol dose of HPIV3/EboGP conferred 100% protection to macaques against EBOV challenge We performed (with JHU clinical collaborators) an open label phase 1 clinical trial to determine the safety, tolerability, and immunogenicity of HPIV3/EboGP delivered IN in healthy adults (NCT025645750), which was intended to be a safety study prior to evaluating aerosol delivery. Ten subjects received 2 doses (4- to 8-week interval) of 6.0 log10 PFU of vaccine. The first dose was moderately infectious (7/10 subjects shed virus detected by qRT-PCR, mean peak titer 3.8 log10 genomic equivalents/ml, mean duration of shedding 7.9 days). Little shedding was detected after the 2nd dose. A second cohort (n=20) received 1 of 2 planned doses of 7.0 log10 PFU of vaccine. Shedding was similar but of shorter duration (mean of 3.7 days). The vaccine was well tolerated, with the exception that asymptomatic ALT elevations were noted in 5 volunteers (3 mild, 2 moderate) in cohort 2 after vaccination and associated with shedding. All resolved by day 28. The study was halted due to these elevations of ALTs, but their significance is unclear. Because of this, this vaccine will not be administered further at this time. Induction of serum antibodies was poor (mucosal antibodies not yet analyzed), but this was expected since, as noted above, we had previously observed that administration by the IN route alone was poorly immunogenic in rhesus monkeys. We also previously developed a 2nd generation version of HPIV3/EboGP in which the HPIV3 F and HN genes were deleted, leaving EBOV GP as the sole viral surface glycoprotein. A large comparative study in cynomolgus monkeys by our collaborator Alexander Bukreyev at the University of Texas Medical Branch, Galveston, (who made the construct while a Staff Scientist in this laboratory) showed that this 2nd generation version is even more protective than the 1st generation even though it is very highly restricted for replication (much more than the 1st generation construct). We have manufactured clinical trial material of this construct for evaluation in adult volunteers. HPIV1 vector: We previously developed a series of attenuating mutations for human parainfluenza virus type 1 (HPIV1) as part of our program to develop live-attenuated intranasal pediatric HPIV1 vaccines. One of these mutations was a 6-nucleotide deletion in the P/C ORF (C170) that impaired the ability of HPIV1 to inhibit host interferon and apoptosis responses. We used HPIV1 bearing this attenuating mutation as a vector to express a codon-optimized version of the EBOV GP gene, which was inserted either at HPIV1 gene position 1, preceding the N gene (pre-N), or at HPIV1 gene position 2, between the N and P genes (N-P). In addition, EBOV GP was expressed either as the full-length protein or as an engineered chimeric form in which its transmembrane and cytoplasmic tail (TMCT) domains were substituted with those of the HPIV1 F protein in an effort to enhance packaging into the vector particle and increase immunogenicity. The four resulting constructs grew to high titers in cell culture and efficiently and stably expressed EBOV GP. When administered to African green monkeys by the combined intranasal (IN) and intratracheal (IT) routes, the HPIV1 constructs were attenuated, replicating at low titers over several days. They were substantially more attenuated than the HPIV3/EboGP construct (see HPIV3 vector, above) administered in parallel, which did not contain specific attenuating mutations. The presence of the EBOV GP gene in the HPIV1 backbone increased its attenuation in addition to the C170 mutation. A single dose of the various HPIV1 constructs was poorly immunogenic. Two doses of the HPIV1 candidates expressing GP from the pre-N position elicited higher GP neutralizing serum antibody titers than the N-P candidates, and the TMCT modification did not increase immunogenicity. Even though the HPIV1 constructs were much more attenuated than the HPIV3/EboGP construct included as a positive control, the titers of EBOV-neutralizing serum antibodies achieved following the second dose of the pre-N HPIV1 constructs were similar to that of HPIV3/EboGP. Unmodified EBOV GP was packaged in substantial quantities into the HPIV1 particle. Replacement of the EBOV GP TMCT with that of the HPIV1 vector F protein did not increase packaging. Evaluation of the stability of expression of the EBOV GP protein following in vivo replication by a double-staining immuno-plaque assay showed that expression of the unmodified form of GP was relatively stable, while that of the TMCT forms was substantially less stable. In conclusion, we identified an attenuated and immunogenic HPIV1-based vaccine candidate expressing EBOV GP from the pre-N position. Because of its substantial level of attenuation, this construct is expected to be well-tolerated in humans and is available for clinical evaluation.